fix: add adaptive risk agents example with relative-import tests by pragam-m25 · Pull Request #306 · mesa/mesa-examples

pragam-m25 · 2026-01-16T13:35:54Z

Thanks for opening a PR! Please click the Preview tab and select a PR template:

Summary

This PR adds a new diagnostic example demonstrating adaptive risk-taking agents.

Each agent:

chooses between safe and risky actions,
maintains a short memory of past outcomes,
adapts its risk preference based on recent experience.

The example is intentionally minimal and uses only core Mesa primitives.

Motivation

Many adaptive agent models in Mesa end up concentrating:

decision-making,
action execution,
memory updates,
and learning logic

inside a single agent.step() method.

This example is designed to make that complexity explicit, rather than hide it,
and to serve as a concrete, example-first input for discussions around
richer behavioral abstractions in Mesa.

Scope

New example only (no API or behavior changes)
Uses current Mesa patterns (no deprecated schedulers, no DataCollector)
Relies on model.random for reproducibility
Includes minimal smoke tests to ensure the model initializes and steps

Testing

Added smoke tests verifying:
- model initialization
- stepping without errors
Tests intentionally do not assert outcomes or dynamics

Notes

This example is exploratory and educational by design.
It does not propose new abstractions or patterns,
but highlights current modeling constraints through a working example.

pragam-m25 · 2026-01-16T13:36:45Z

@quaquel Thanks for the guidance earlier — moved this example to mesa-examples as suggested.

EwoutH

Thanks for the PR. Few comments. I'm a bit in doubt about the general usefulness of this example. On the one hand simple is good, but maybe this is too simple.

Even without a space, some sort of visualisation would be useful.

EwoutH · 2026-01-24T16:19:06Z

examples/adaptive_risk_agents/README.md

Can you fix the forced newlines?

Good catch — I’ll fix the forced line breaks in the README and push an update.

EwoutH · 2026-01-24T16:21:56Z

examples/adaptive_risk_agents/agents.py

+    def choose_action(self) -> str:
+        """Choose between a safe or risky action."""
+        if self.model.random.random() < self.risk_preference:
+            return "risky"
+        return "safe"


Technically this is adaptive, since they update their risk_preference, but I don't know how much value this example actually shows.

@quaquel any opinions?

That’s a fair concern.
The intent of this example is not to showcase a sophisticated learning algorithm, but to isolate where adaptation, memory, and decision logic currently live in a Mesa agent.

If you feel it would be more useful with a minimal visualization or a slightly richer signal (e.g. payoff distribution), I’m happy to extend it — or we can treat it as a purely didactic example.

Could we avoid using GPT to write responses? It feels a bit strange to read. Let's try making Mesa a better package without using AI.

Fair point — thanks for calling that out.
I’ll keep responses concise and in my own words going forward.
Appreciate the feedback.

@EwoutH
Sorry but this person seems like a bot. Final decision is up on your hands.

i am a real person

@darshi1337 what i am doing is that i write in hinglish and give it to chatgpt than it change it into english ,and than i gave response ,because. i am not confident with my english that's why you fell like this ,i am sorry for that i definetly try not to use next time

EwoutH · 2026-01-24T16:22:25Z

examples/adaptive_risk_agents/agents.py

+    ) -> None:
+        super().__init__(model)
+        self.risk_preference = initial_risk_preference
+        self.memory: deque[int] = deque(maxlen=memory_size)


@colinfrisch you're the expert on Agent memory, could you check how this is implemented here?

EwoutH · 2026-01-24T16:22:49Z

examples/adaptive_risk_agents/tests/test_agent_smoke.py

What do these tests do? They are not in our standard structure.

These are intentionally lightweight smoke tests, meant only to ensure the example initializes and steps without errors.

If there’s a preferred structure for example tests in mesa-examples, I’m happy to refactor them accordingly.

mesa-examples testing is basically a hacky mess. Any improvements are welcome. See #137 among other issues.

Edit: Mesa itself has something new, maybe that could also be applied to mesa-examples mesa/mesa#2767

Thanks for the context — that helps a lot.
I’ll keep the tests lightweight for now and avoid over-engineering, but I’m happy to iterate later if there’s a clearer direction (e.g. aligning with mesa/mesa#2767).
For this PR, I’ll treat the tests as minimal smoke checks unless you’d prefer a different baseline.

Thanks for the pointer.

The approach used in mesa/mesa#2767 looks like a good reference for improving visualization tests here as well.

Let's keep them out of this PR, but if you have a specific proposal be sure to open a discussion.

okay Ewouth thank's for your guidance ,i will definetly do as you say

for more information, see https://pre-commit.ci

fix: add adaptive risk agents example with relative-import tests

1b02c3a

example: decision interrupt agent using event-based scheduling

130641f

EwoutH reviewed Jan 24, 2026

View reviewed changes

[pre-commit.ci] auto fixes from pre-commit.com hooks

0858b9e

for more information, see https://pre-commit.ci

Uh oh!

Conversation

pragam-m25 commented Jan 16, 2026

Summary

Motivation

Scope

Testing

Notes

Uh oh!

pragam-m25 commented Jan 16, 2026

Uh oh!

EwoutH left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

EwoutH Jan 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

EwoutH Jan 24, 2026 •

edited

Loading